PROJECT 1: NEUROIMAGING OF HIPPOCAMPAL SUBFIELDS IN OLDER ADULTS & MCI A major goal for health research in the U.S. is to discover new disease-modifying therapies and target individuals susceptible to AD as early as possible. Episodic memory deficits and changes in the medial temporal lobes have been highlighted both in healthy aging and for their potential as early markers of AD. We propose to collect a comprehensive set of behavioral and high-resolution neuroimaging data from (1) healthy elderly individuals, (2) elderly individuals with diagnosed amnestic mild cognitive impairment (aMCI), and (3) elderly individuals with milder impairments who do not reach criteria for an MCI diagnosis, but are nevertheless impaired on memory tests (amnetic, early impairment - aEI). We will use these data to test several key predictions of a neurocognitive model of age-related memory impairment. This work is based on converging insights from computational models as well as behavioral, electrophysiological, and neuroanatomical findings in rodent models of aging that place emphasis on the role of the dentate gyrus in pattern separation (the ability to isolate similar memories from each other). The model posits that memory impairments in the course of aging and early dementia are due to the degradation in input to the dentate gyrus and CAS region from layer II entorhinal cortex neurons, which leaves the system with an impaired ability to perform pattern separation. We propose to test predictions of this model using targeted behavioral assays sensitive to hippocampal and perforant path integrity as well as cutting-edge neuroimaging techniques (high-resolution functional and structural MRI, resting state functional connectivity, ultrahigh-resolution microstructural diffusion tensor imaging). We predict that compared to healthy elderly, those with memory impairments will have behavioral deficits consistent with reductions in pattern separation and that these are tied to structural and functional changes in the dentate gyrus, the CA3, and their perforant path input. We will also investigate the modulatory effect of individual differences in our healthy elderly group, as well as the potential effect of ApoE4 genetic susceptibility. This rich dataset will also be useful beyond our hypotheses and questions. Thus, we will work with the ADRC's Data Core to generate new hypotheses, accurately quantify our outcome measures in a reliable way, and conduct statistically rigorous analyses that combine multiple imaging modalities with behavioral and neuropsychological data in order to isolate useful biomarkers that accurately discrminate among groups and potentially predict conversion from healthy aging to dementia.